Is an OWL-DL property considered a Binary Predicate or a Unary Predicate?

classic Classic list List threaded Threaded
3 messages Options
Reply | Threaded
Open this post in threaded view
|

Is an OWL-DL property considered a Binary Predicate or a Unary Predicate?

brian dorsey
Hi all,

How does one view a property in OWL 1.0 (SHOIN(D))?

Is it a "binary predicate" or is it defined as a kind of concept (unary predicate)?

My presumption is that roles/object-properties in OWL-DL are binary predicates.

If I have the following:

hasProp(x,y)

where x and y are individuals of ClassX and ClassY respectively.

I can see how this can be seen asĀ  a binary predicate, particularly when I think of this as an A-Box assertion.

However, if I have the following T-Box assertion

\exists_{\geq 1} hasProp.ClassY


I can see how this can be seen as an unary predicate (perhaps I am wrong), in that this statement represents the anonymous class of individuals that hold the "hasProp" relation with individuals in ClassY, for example individual x.



[ +++
My understanding is that Protege 3.4 uses OWL 1.0, and not the later OWL2.
I am not sure if properties can be both in OWL2 or if it makes any difference what version of OWL-DL is used.
+++]


regards,
Brian

_______________________________________________
protege-owl mailing list
[hidden email]
https://mailman.stanford.edu/mailman/listinfo/protege-owl

Instructions for unsubscribing: http://protege.stanford.edu/doc/faq.html#01a.03
Reply | Threaded
Open this post in threaded view
|

Re: Is an OWL-DL property considered a Binary Predicate or a Unary Predicate?

Thomas Russ
Short summary:
   Classes are unary predicates.
   Properties are binary predicates.

On Feb 25, 2011, at 1:04 AM, brian dorsey wrote:

> Hi all,
>
> How does one view a property in OWL 1.0 (SHOIN(D))?
>
> Is it a "binary predicate" or is it defined as a kind of concept  
> (unary predicate)?
>
> My presumption is that roles/object-properties in OWL-DL are binary  
> predicates.

Correct.

>
> If I have the following:
>
> hasProp(x,y)
>
> where x and y are individuals of ClassX and ClassY respectively.
>
> I can see how this can be seen as  a binary predicate, particularly  
> when I think of this as an A-Box assertion.
>
> However, if I have the following T-Box assertion
>
> \exists_{\geq 1} hasProp.ClassY
>
> I can see how this can be seen as an unary predicate (perhaps I am  
> wrong), in that this statement represents the anonymous class of  
> individuals that hold the "hasProp" relation with individuals in  
> ClassY, for example individual x.

That is correct.  This is a unary predicate because it is a class  
description.  It happens to be a defined class, but all classes are  
unary predicates.  The fact that the description references a binary  
predicate as part of the description doesn't change anything.

> [ +++
> My understanding is that Protege 3.4 uses OWL 1.0, and not the later  
> OWL2.
> I am not sure if properties can be both in OWL2 or if it makes any  
> difference what version of OWL-DL is used.
> +++]

It doesn't matter which version of OWL is used.  The same principles  
apply.

The only matter that is at issue is whether a Class or Property can  
also be and OWL individual.  In OWL 1, this is only possible in the  
OWL-Full version of OWL.  In OWL 2, you can also have something like  
OWL Full, but it is also permitted to use "punning" to have an  
individual with the same identifier as a class or property.


_______________________________________________
protege-owl mailing list
[hidden email]
https://mailman.stanford.edu/mailman/listinfo/protege-owl

Instructions for unsubscribing: http://protege.stanford.edu/doc/faq.html#01a.03
Reply | Threaded
Open this post in threaded view
|

Re: Is an OWL-DL property considered a Binary Predicate or a Unary Predicate?

brian dorsey
Thanks Thomas.


_______________________________________________
protege-owl mailing list
[hidden email]
https://mailman.stanford.edu/mailman/listinfo/protege-owl

Instructions for unsubscribing: http://protege.stanford.edu/doc/faq.html#01a.03